Audio AI
4 stories in this blend
There's An AI For That·5d agoResearchers leverage artificial intelligence to analyze animal vocalizations
Scientists are training machine learning models on vast audio datasets to interact directly with species such as birds, elephants, and sperm whales. In one trial, an AI model generated original finch calls that prompted real birds to respond in real time. The primary challenge remains accurately mapping sound patterns to specific meanings.
Superhuman·Oct 14Fish Audio releases Drama 3 text to speech voice generator
Fish Audio unveiled Drama 3, an expressive speech model allowing users to control word pronunciation and emotional delivery through text descriptions. The platform also enables targeted audio updates without regenerating entire audio tracks.
meet cue·Sep 29ElevenLabs Releases Multi Language Expressive Speech Models
ElevenLabs announced the v4 edition of its audio generation tool, featuring lower latency and support for over 90 international languages. The update allows users to clone target voices from just ten seconds of sample audio.
Superhuman·Sep 7Google brings Lyria 3.5 music generation model directly into Gemini and developer APIs
Google has updated its Gemini platform with Lyria 3.5, allowing users to compose full musical tracks up to three minutes long. The upgraded audio engine features clearer melodies and richer instrumental arrangements for creative projects.
